1. Field of Invention
The present invention relates to a method for filtering data, and more particularly to a method for filtering file clusters.
2. Related Art
Nowadays due to the rapid growth of the amount of data, it becomes a problem in urgent need of solutions for modern people to find required files from a large number of data files. Therefore, many manufacturers provide file search software, through which not only the desired data can be found at a local end, but also the required data files can be found in the network.
Currently, a search engine in the network may also achieve a function similar to advanced filter. However, in each action, a keyword needs to be input again to obtain data from a server of the search engine, resulting in a waste of time. For example, in a well-constructed search engine in the network, it is assumed that 100 search results are obtained by inputting “NBA”. An advanced search function of the search engine is used to filter the obtained results by further inputting an advanced filter keyword in the 100 search results of “NBA”, for example, “Yao-ming” is further input to obtain 50 filtered results through the advanced search. However, at this time, the previous 100 search results of “NBA” are already cleared and only 50 results obtained through the advanced search of “Yao-ming” are left. If a user intends to search other keywords from the 100 search results of “NBA”, the user needs to input “NBA” again to obtain new search results before making the advanced search once again. As such, pages and data of the found results cannot be stored, and data blocks cannot be divided, so that it is impossible for the user to select a single data block for the advanced filter action.
Besides, the most widely used Google search engine is unable to change main search data blocks, and thus the Google search engine can only perform a further search. For example, “NBA” is input in the Google search engine and it is assumed that 100 search results are obtained. If data containing “Yao-ming” needs to be obtained from the 100 results, keywords “NBA Yao-ming” must be input in the Google search engine again to obtain data of “NBA” plus “Yao-ming”.
Moreover, most of the network search engines provide an “all these words” function similar to filter in advanced search. However, the filtered results do not conform to the filter conditions completely. For example, a combination of three keywords “NBA Coby Laker” is typed in the field of “all these words” for search. The found results include all data that partially and completely conforms to the filter conditions. For example, a plurality of data that merely conforms to the parts such as “NBA Coby”, “NBA Laker”, and “Coby Laker” but is not the target of the advanced filter is found, instead of the desired data that completely conforms to “NBA Coby Laker”. Therefore, the advanced filter lacks in accuracy and the found data is more disordered.
In addition, currently a user is also unable to decide new filter conditions from a filter history. For example, first-layer filter results with “NBA” are used for illustration. When the user inputs a new filter condition “Coby”, search results containing “Coby” are filtered from the results of “NBA”. If the results of “Laker” are filtered again with “NBA”, the original search results of “Coby” are cleared.
In an existing widely applied website search engine (for example, Google, Yahoo, and MSN), upper-layer filter results can be backtracked only once. If filter with different filter conditions is performed on the upper-layer filter results, the original filter conditions are refreshed.